This invention relates to the synthesis, deprotection, and purification of molecules comprising one or more ribonucleotides.
The following discussion relates to the synthesis, deprotection, and purification of oligonucleotides containing one or more ribonucleotides. The discussion is not meant to be complete and is provided only for understanding the invention that follows. The summary is not an admission that any of the work described below is prior art to the claimed invention.
Research in the many roles of ribonucleic acids has, in the past, been hindered by limited means of producing such biologically relevant molecules (Cech, 1992, Nucleic Acids Research, 17, 7381-7393; Francklyn and Schimmel, 1989, Nature, 337, 478-481; Cook et al., 1991, Nucleic Acids Research, 19, 1577-1583; Gold, 1988, Annu. Rev. Biochemistry, 57, 199-233). Although enzymatic methods existed, protocols that allowed one to probe structure function relationships were limited. Only uniform post-synthetic chemical modification (Karaoglu and Thurlow, 1991, Nucleic Acids Research, 19, 5293-5300) or site directed mutagenesis (Johnson and Benkovic, 1990, The Enzymes, Vol. 19, Sigman and Boyer, eds., 159-211) were available. In the latter case, researchers were limited to usage of natural bases. Fortunately, adaptation of the phosphoramidite protocol for DNA synthesis to RNA synthesis has greatly accelerated our understanding of RNA. Site-specific introduction of modified nucleotides to any position in a given RNA has now become routine. Furthermore, one is not confined to a single modification but can include many variations in each molecule.
It is seemingly out of proportion that one small structural modification could cause such a dilemma. However, the presence of a single hydroxyl at the 2′-position of the ribofuranose ring, has been the major reason that research in the RNA field has lagged so far behind comparable DNA studies. Progress has been made in improving methods for DNA synthesis that have enabled the production of large amounts of antisense deoxyoligonucleotides for structural and therapeutic applications. Only recently have similar gains been achieved for ribonucleotides (Wincott et al., 1995, Nucleic Acids Research, 23, 2677-2684; Sproat et al., 1995, Nucleosides and Nucleotides, 14, 255-273; Vargeese et al., 1998, Nucleic Acids Research, 26, 1046-1050).
The chasm between DNA and RNA synthesis is due to the difficulty of identifying orthogonal protecting groups for the 5′- and 2′-hydroxyls. Historically, two standard approaches have been taken by scientists attempting to solve the RNA synthesis problem; developing a method that is compatible with state-of the-art DNA synthesis or designing an approach specifically suited for RNA. Although adaptation of the DNA process provides a more universal procedure in which non-RNA phosphoramidites can easily be incorporated into RNA oligomers, the advantage to the latter approach is that one can develop a process that is best for RNA synthesis and as a result, better yields can be realized. However, in both cases similar issues are faced, for example identifying protecting groups that are compatible with synthesis conditions yet can be removed at the appropriate juncture. This problem does not refer only to the 2′- and 5′-OH groups, but includes the base and phosphate protecting groups as well. Consequently, the accompanying deprotection steps, in addition to the choice of ancillary agents, are impacted. Another shared issue is the need for efficient synthesis of the monomer building blocks.
Solid phase synthesis of oligoribonucleotides follows the same pathway as DNA synthesis. A solid support with an attached nucleoside is subjected to removal of the protecting group on the 5′-hydroxyl. The incoming phosphoramidite is coupled to the growing chain in the presence of an activator. Any unreacted 5′-hydroxyl is capped and the phosphite triester is then oxidized to provide the desired phosphotriester linkage. The process is then repeated until an oligomer of the desired length results. The actual reagents used may vary according to the 5′- and 2′-protecting groups. Other ancillary reagents may also differ.
Once the oligoribonucleotide has been synthesized, it must then be deprotected. This is typically a two-step process that entails cleavage of the oligomer from the support and deprotection of the base and phosphate blocking groups, followed by removal of the 2′-protecting groups. Occasionally, a different order of reactions or separate deprotection of the phosphate groups is required. In all cases, it is imperative that indiscriminate removal of protecting groups not occur, this is particularly an issue in the classic situation wherein the first step is base mediated. In this case, if the 2′-hydroxyl is revealed under these conditions, strand scission will result due to attack of the vicinal hydroxyl group on the neighboring phosphate backbone. Two other concerns that are prevalent in RNA synthesis but play no part in DNA are the propensity for 3′-2′ phosphodiester migration to provide undesired 2′-5′ linkages and the susceptibility of oligoribonucleotides to degradation by ribonucleases. The latter fact has led many researchers to develop 2′-protecting groups that can remain in place until the oligomer is required for the desired experiment.
In the past, deprotection of oligoribonucleotides containing a 2′-O-TBDMS (t-butyldimethylsilyl) group was a two step process that first entailed a basic step similar to that used for the deprotection of DNA in which the oligomer was cleaved from the support and the base and phosphate groups were removed. The initial step was accomplished in 1-4 h at 55° C. with 3/1 NH4OH/EtOH. Since the oligomer is not exposed to severe deprotection conditions for prolonged periods, better yields of higher quality product result. More recently, a faster, two step, deprotection protocol, entailing the use of aqueous methylamine has been reported for RNA (Usman et al., U.S. Pat. No. 5,804,683; Wincott et al., 1995, supra; Reddy et al., 1995, Tetrahedron Lett., 36, 8929-8932). Incubation times have been reduced to 10 min at 65° C. When compared with other RNA deprotection methods, treatment with this reagent gave greater full length product than the standard protocol using 3/1 NH4OH/EtOH (Wincott et al., 1995, supra). The only requirement is that acetyl must be used as the N-protecting group for cytidine because of a well-documented transamination reaction (Reddy et al., 1994, Tetrahedron Lett., 35, 4311-4314). As stated earlier, through the use of methylamine this step has been reduced to 10 minutes. The second step is removal of the 2′-silyl protecting group from the oligonucleotide. In the past this had been accomplished with 1 M n-tetrabutyl ammonium fluoride (TBAF) in THF at room temperature over 24 h (Usman et al., 1987, J. Am. Chem. Soc., 109, 7845-7854; Scaringe et al., 1990, Nucleic Acids Research, 18, 5433-5341). Unfortunately, the use of this deprotecting agent produces salts which must be removed prior to analysis and purification. In addition, the long exposure time required for complete removal of the protecting group, coupled with the reagent's sensitivity to adventitious water (Hogrefe et al., 1994, Nucleic Acids Research, 21, 4739-4741), made it a less than ideal reagent. Although some reports have been published regarding the use of neat triethylamine trihydrofluoride (TEA.3HF) (Duplaa et al., U.S. Pat. No. 5,552,539, Gasparutto et al., 1992, Nucleic Acids Research, 20, 5159-5166; Westman et al., 1994, Nucleic Acids Research, 22, 2430-2431) as a desilylating reagent, results have been mixed. A cocktail of TEA.3HF in combination with N-methyl pyrrolidinone (NMP) (Usman and Wincott, U.S. Pat. No. 5,831,071; Wincott et al., 1995, supra) or DMF (Sproat et al., 1995, supra) has also been described in which full deprotection can be achieved in 30-90 min at 65° C. or 4-8 h at room temperature. As an added advantage, since no salts are produced, the product can be directly precipitated from the desilylating reagent.
Tracz, U.S. Pat. No. 5,977,343; Tracz, U.S. Pat. No. 5,686,599, describes a one-pot protocol for ribonucleotide deprotection using anhydrous methylamine and triethylamine trihydrogen fluoride. This procedure involves the use of anhydrous methylamine followed by neat triethylamine trihydrofluoride to effectively deprotect oligoribonucleotides in a one-pot fashion. However such a protocol may be cumbersome for deprotection of oligonucleotides synthesized on a plate format, such as a 96-well plate, because it may be necessary to separate the solid-support from the partially deprotected oligonucleotide prior to the 2′-hydroxyl deprotection. Also, since the methylamine solution used is anhydrous, it may be difficult to solubilize the negatively charged oligoribonucleotides obtained after basic treatment. More recently this procedure has been reported in which both the basic deprotection and the desilylation reaction can be accomplished in one-pot using a mixture of anhydrous methylamine in ethanol followed by addition of TEA-3HF (Bellon, 1999, Current Protocols in Nucleic Acid Chemistry, Beaucage, Bergstrom, Glick and Jones, eds., in press). This protocol allows for the complete deprotection of an oligoribonucleotide in less than 2 h without any evidence of 3′-2′ migration.
The parameters of 2′-deprotection are dictated by the corresponding protecting groups utilized for differing 2′-chemistries present within a given oligonucleotide. The use of alternate 2′-ribofuranosyl carbocycle functions within the same oligonucleotide molecule can present potential problems with respect to the synthesis, deprotection, and purification of such molecules. The efficient synthesis of nucleic acids which are chemically modified to increase nuclease resistance while maintaining catalytic activity is of importance to the potential development of new therapeutic agents. Recently, Beaudry et al., 2000, Chemistry and Biology, 7, in press, describe the in vitro selection of a novel nuclease-resistant RNA phosphodiesterase. This enzymatic nucleic acid molecule can contain both ribo (2′-hydroxyl) and amino (2′-deoxy-2′-amino) functions. The large scale synthesis of oligonucleotides with both ribo and amino functions presents practical problems with regard to the concomitant removal of tert-Butyldimethylsilyl (TBDMSi) and N-phthaloyl protecting groups, while at the same time preserving the integrity of the ribonucleotide linkages. The use of the N-phthaloyl protecting group for the 2′-amino group during oligonucleotide synthesis offers the benefit of improved synthetic yields compared to the trifluoroacetyl (TFA) and FMOC groups (Usman et al., U.S. Pat. No. 5,631,360; Beigelman et al., 1995, Nucleic Acids Research, 23(21), 4434-4442). The phthaloyl group is readily cleaved with aqueous methylamine at 65° C. and the TBDMSi group is readily cleaved using a fluoride ion source, such as tetrabutylammonium fluoride (TBAF) or triethylammonium trihydrofluoride (TEA.3HF). Application of the “one pot” deprotection procedures described above results in the incomplete deprotection of N-phthaloyl protection. The two step deprotection procedure can be employed for the complete deprotection of oligonucleotides containing both ribo (2′-TBDMS) and amino (N-phthaloyl) protecting groups, however, this process is not readily amenable to large scale oligonucleotide synthesis or multiwell plate oligonucleotide synthesis.
As such there exists an unmet need for a fast, efficient method which allows for the complete deprotection of molecules containing both amino and ribo carbohydrate moieties. Such a method will enable the large scale synthesis of such molecules for use as therapeutic agents and the small scale synthesis of such molecules for combinatorial screening.